ad-creative
Pass
Audited by Gen Agent Trust Hub on Jul 14, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill processes untrusted performance data (such as CSV files, text pastes, or outputs from ad platform APIs) to iterate on creative variations. This represents an indirect prompt injection attack surface where malicious instructions could be hidden within the data to influence the agent's behavior.
- Ingestion points: Untrusted data enters the agent context through performance metrics provided by the user or fetched via tools (e.g.,
google-ads reports get). - Boundary markers: The skill lacks explicit delimiters or instructions to the agent to ignore commands embedded within the performance data.
- Capability inventory: The agent has the ability to perform network operations and manage campaigns via ad platform tools (
google-ads,meta-ads, etc.) and read local marketing context files. - Sanitization: The instructions do not specify any validation or sanitization for the performance data before it is processed by the agent.
Audit Metadata